Currently available surgical trackers are inadequate for microsurgery, and especially for robotic microsurgical systems where the tracker is inside the control loop. This problem is exacerbated in the case of handheld microsurgical systems, where small and rapid (1-20 Hz) disturbances such as physiological tremor must be correctly tracked and accounted for by the control system. As a result, for the most demanding types of microsurgery, such as vitreoretinal microsurgery, few suitable tracking systems exist. Available commercial systems are not designed for microsurgery, and their accuracy is inadequate for the task. More importantly, their latency makes them unsuitable for inclusion within the control loop of active handheld robotic instruments. The two main operating principles used in surgical trackers are optical and electromagnetic (EM). Optical trackers have better accuracy, but have a serious flaw when used for closed-loop control: any obstruction of sensor sightlines may cause control to fail. The surgeons who have used ?Micron,? a handheld micromanipulator under development in our laboratory, have repeatedly identified this need for sightlines as the main factor preventing it from being clinically compatible. Micron is just one example; any handheld instrument incorporating closed-loop control and optical tracking (e.g., Navio?) faces the same problem. We propose to develop an innovative medical tool tracker, the In-Loop Electromagnetic Tracker (ILEMT). The goal: > 50x improvement in latency, while maintaining the interference rejection of the best existing EMTs. Dual-path high/low carrier modulation combines the best qualities of AC and pulsed-DC EMTs, while frequency-domain multiplexing pushes bandwidth and resolution well beyond that of commercial offerings. A class-D current source will efficiently drive arbitrary signals into the source coil, allowing simultaneous emission of two or more carriers on each axis for metallic error compensation, with low-noise signal conditioning and data acquisition will to maximize resolution. A software-defined architecture implements signal processing in software and FPGA firmware, maximizing flexibility. The ultimate goal of this research is to develop a production-quality EMT with an open architecture (under CC-BY copyright), as a crucial enabling technology for handheld microsurgical robotics. It also has potential for significant advances in the science of microsurgery, and for microsurgical assessment in vivo. The specific aims are to: 1. Develop a three-channel prototype and demonstrate measurement in six degrees of freedom (6DOF). 2. Develop signal processing algorithms for metallic and EM interference rejection. 3. Develop a reference ILEMT design and test it to verify performance goals are achieved. We will compare ILEMT performance with 2 commercial EM trackers. 4. Integrate ILEMT with the instruments Micron and Navio and test in animals in vivo and in cadavers.